[ExI] extropy-chat Digest, Vol 265, Issue 62

[Jason Resch]

On Fri, Oct 31, 2025, 4:12 AM Ben Zaiboc via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> On 31/10/2025 00:38, bill w wrote:
>
> I have read several times in these chats the assumption that one cannot understand something as complicated as themselves.
> Why not?  It sounds reasonable but what's the basis for it?   bill w
>
>
> This is a red herring. The main problem with it is that we have no agreed
> common meaning for the word 'understand', so different people can interpret
> it differently.
>


Agreed, I was clear to specify *fully understand* and "everything there is
to know about a human brain." I will say more about this below.


> Jason has gone into some details of an information-theory view, which is
> fine, but hardly applies to real-world scenarios. I think the main issue is
> not what is theoretically possible, but what do we need?
>

I'll rank degrees of understanding for varying purposes:

Low level understanding:
- Knowing about neurons and their disease pathologies
- Finding medicines to treat biological deficiencies of neurons,
neurotransmitters, brain metabolism, etc.
- How neurons behave and react under different conditions

Mid level understanding:
- Knowing the algorithms brain regions and cortical columns employ to solve
various problems
- Knowing the inputs and outputs of this processing for the millions of
cortical columns and hundreds of brain regions
- Knowing the meaning of the messages/information exchanged and passed
between brain regions


High level architectural understanding:
- Knowing what deficits correspond to brain damage to different areas
- Planning a brain surgery to be minimally disruptive
- Knowing which brain regions are connected to which others and the
information flows between them

Top level (Mental/Psyche) understanding:
- Knowing the person's general personality traits, dispositions,
preferences, abilities, etc.
- Being able to model with some accuracy, how a person might behave in
different circumstances
- How to best tweak or nudge the person via various psychiatric forms of
treatment, towards some productive change in behavior or personality.

Full/complete understanding:
- Predicting treatment outcomes with perfect as accuracy
- Predicting behavior with perfect accuracy


What the word "understand" means to me, is knowing something well enough to
be able to model it and make predictions about how the thing in question
will respond in different imagined scenarios. The better the degree of
understanding, the higher the degree of predictive accuracy. Perfect
prediction requires a complete or full understanding, of both the object
and the environment (for the particular scenario).


> Individually, we hardly understand anything at all. How many car drivers
> actually have much of an understanding of the car they drive? But they all
> have an understanding of what they need to know in order to drive.
>

Right new should not confuse "understanding how to drive" with
"understanding how their car works"

Even mechanics, while knowing how cars generally work, have to look up
manufacturer specific guides for how to perform different operations on
different vehicles. And even then, we can't expect them to understand the
thermodynamic details of the combustion reaction, the molecular chemistry
of the tires, or the software of the GPS navigation system.


> I think I understand how my computer works. Which is laughable if you take
> into account all the things that make up a computer, both hardware and
> software. I hardly understand a tiny fraction of it all. But I can still
> build one from parts I can buy, install an operating system and application
> software and do useful things with it. I understand enough. Computers are
> enormously complicated, but people, collectively, have created them. No
> single person could do it. No single person understands everyting about a
> computer. But we have them.
>

It really is quite miraculous when you think about it.


> So I think the important thing is not "understand" something, but
> "Understand x about" something, to some practical end involving x. The
> question is: What do you want to do? Then you can decide what you need to
> understand.
>

Yes, I don't mean to discourage attempts at research because it ball adds
to the collective understanding of human civilization, even though no one
person or brain can read every book and hold that collective knowledge in
their head (we can at least look up the part we need to know to achieve the
goal we want).

It is perhaps a bit like a CPU in a computer. A CPU can only hold a handful
of register values in its "head" at a time. Nevertheless, by cycling values
in and out, and offloading them to eternal memory/storage, there is no
limit to what a CPU is able to compute (or do).

So we might think of ourselves as individual CPUs, accessing the memory we
need, loading it into our heads, and performing a small discrete task that
inches is closer to a goal.

Jason
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20251031/f18f506f/attachment.htm>